Univariate and Multivariable analysis of predictors of malaria infections.

*Multivariable polynomial fraction showed age has a non linear effect in all the cohorts (see Figure S3),
#: The best fit model was obtained with only first nearest distances in the model,
§: Risk ratios are for each step increase in 0.35 and 0.45 power function of distance to the infected and uninfected child respectively.

* 多变量多项式分数（Multivariable polynomial fraction）分析显示，所有队列中年龄均呈现非线性效应（详见补充图S3）；
# 最优拟合模型仅纳入一阶最近距离变量；
§ 风险比分别对应与感染儿童和未感染儿童的距离的0.35次幂、0.45次幂每增加一个步长的结果。